Design a recommendation homepage
Medium45 minFree, no account
No query, no intent, and a cold-start problem on both sides.
The question
Design the recommendation system behind a streaming service homepage: rows of titles, personalised, with no query to work from.
200M users, 100k titles.
Functional
- Several themed rows, each personalised and ordered.
- Reasonable recommendations for a brand-new user.
- New titles get exposure.
Non-functional
- Page assembled in under 300ms.
- Reflect what the user watched an hour ago.
45:00Commit to an answer before you open the solution. Reading it first teaches you to recognise good answers, which is not the skill being tested.
Stuck?
0 of 3 hints takenThe worked solution
written by a person · not a gradeScore yourself
0 of 5 marked- Separated candidate generation from ranking with a latency reason20
- Multiple complementary retrieval sources, including non-ML ones20
- Two-tower embeddings with negative sampling addressed20
- Both cold starts handled with concrete fallbacks25
- Treated row selection and diversity as a problem15
We run no AI here and nothing on this page grades you. The score is yours, and the useful number is the one you get on the same problem a month from now, cold.
kept in this browser only